How much wood would a woodchuck chuck if a woodchuck could chuck wood? As much forex as the RBI would hoard in Mother Hubbard’s cupboard?

We just don’t know, and that, surely, ought to be a problem worth considering.

In fact, it is impossible not to draw a comparison with the food stocks problem we are currently facing. For instance, in 1983, the late Raj Krishna and Ajay Chibber wrote a paper for IFPRI in which they estimated that India needed no more than 3-4 million tonne of buffer stock.

This calculation was based on the premise that the reason you needed it was to intervene in the market to influence prices. Today, India has 44 million tonne of foodgrains, and the cost of carrying it is bankrupting the exchequer.

This, indeed, is the reason why it might be worth asking if the same thing might not happen with forex reserves as well. The point, ultimately, is this: can a country have too much foreign exchange reserves? For instance, if by a miracle, India’s foreign exchange reserves doubled tomorrow, would they still be described as ‘adequate’, or would they then be in ‘excess’ of our needs?

Not long ago, reserves worth six months of imports were described as adequate. Now, at nearly $40 billion, when they are much higher at nearly 8 months of imports, they are still described as adequate. This suggests that it is no longer the rule of thumb of 6 months of imports, but some other rule that determines the desired level.

It could be, for instance, what the RBI governor recently referred to, namely, the ‘Guidotti rule’, which suggests that a country must hold enough reserves so that it need not borrow for a year.

But, just like food stocks, holding hard currency assets also involves a cost and it may be possible that, at some point, these costs exceed the benefits from holding them. Thus, if $40 billion is the magic number, then would the reserves always be maintained at that level? And if it is not the magic number, why did we borrow with the MIBs?

It is useful to bear in mind here that forex reserves accrue on account of ‘involuntary’ inflows of dollars to the country on the current and capital account. If imports are higher than exports, the capital account surplus over and above the current account deficit is the amount by which reserves would rise, if no proactive steps are taken to maintain them.

It was such involuntary flows of foodgrains that raised stocks to inoptimal levels. In the case of foreign exchange, it is possible that an increase in exports or inflow of foreign investment could push up reserves.

But if ‘involuntary’ capital flows decline, the government may decide to borrow to push up the level of reserves. How much should it aim to borrow and at what cost? This lies at the heart of the problem, and unfortunately for the central bank, there are no clear rules as to how much. The objective therefore seems to be to simply build reserves till it is ‘felt’ that they are ‘enough’. But what are the feelings based on?

Again, there are no answers. Much also depends on what the reserves are needed for. Clearly, with full capital account convertibility and the currency floating freely, a central bank need hold no foreign exchange reserves at all. But very few countries allow their currency to move about completely freely.

So, when the currency is a managed float, as in the case of the rupee, a central bank might hold reserves so that it can intervene in the foreign exchange market to manipulate the exchange rate. The central bank might intervene either to reduce volatility or to maintain the value of the currency.

This ought to suggest that, depending on the objectives, the volume of reserves to be held may be different. And, just as in the case of food stocks, if the purpose is curb volatility, the desired level may be much lower, since volatility is usually the result of temporary mismatches in demand and supply.

What if the objective of the central bank’s exchange rate policy is to determine the value of the currency? If the macroeconomic fundamentals of a country are weak, it is unlikely that any amount of intervention made possible by large holdings of reserves will be able to prevent depreciation.

Intervention is effective usually only at the margin. The experiences of both the Bank of England and the Bank of Japan indicate that when there are huge pressures on the currency, central bank interventions become ineffective.

Hence, if the government believes that macroeconomic fundamentals are weak, it doesn’t really help if it holds reserves. But it may still want to, hoping that it will be able to defend the currency in the event of a crisis.

A third reason to hold reserves would be the possibility of external exogenous shocks which reserves would help to minimise. But how much would be needed?

Reserves are costly to hold. Usually foreign exchange assets consist of hard currencies. Since hard currencies are considered to have relatively lower risk, the investment of these assets in, for instance, US treasury bills, earns a relatively low return.

Moreover, money that flows into the country on account of the purchase of these reserves are in the weaker domestic currency, the rate of return on them being usually higher to cover the higher risk. The costs of holding reserves will be the difference between the returns the central bank can earn on its hard currency reserves and the returns to foreign capital invested domestically. Clearly, these costs may be quite high and need to be kept in mind while deciding the volume of reserves that a country should hold. One way to reduce the cost may be to try to maximise the returns that may be earned on them.

Thus, instead of holding them in US treasury bills, that bring in among the lowest returns on dollar holding, they may be invested in higher interest earning assets.

The other way would be to minimise the premium India pays on borrowing funds to build reserves. Or, to find the cheapest funds available to the country. In the case of the recent borrowing by India through the Millennium India Development bonds from NRIs, it has been alleged that cheaper loans could have been arranged from multilateral lending agencies.

To go back to the food stocks analogy, it has not been easy to decide how to reduce them. Domestic sales were more or less ruled out as they would affect the price which would upset farmers.

What would be the analogue in the case of foreign exchange assets? That would result in currency appreciation, which would upset exporters. Clearly, it is important to factor in the cost of holding reserves, and decisions to borrow should be part of a well thought out exchange rate policy.

The RBI probably has one. In which case, why not tell everyone? Why create mysteries?